Dynamic hotspots of nitrous oxide and methane in coastal marshes:

Responses to two long-term fertilization experiments

Serena Moseman-Valtierra1, Kevin D. Kroeger2

smoseman@mail.uri.edu 1 University of Rhode Island, Department of Biological Sciences,

Kingston RI 02881 2 USGS Coastal and Marine Science Center, Falmouth MA 02540

http://sdo.gsfc.nasa.gov/

Despite small size, coastal salt marshes affect global carbon sequestration

Temperate Forests: 53 Tg C year-1

Tropical Forests: 78.5 Tg C year-1

Salt Marshes: 5-87 Tg C year-1

Reviewed in Mcleod et al. 2011 Front. Ecol. Environ.

0

-2

-1

0

1

2

3

4

Short-term nitrate addition shifted salt marsh from N2O sink to source

= Nitrate added = Control

July ‘09 April ‘10 June ‘10

(Moseman-Valtierra et al. 2011. Atmospheric Environment)

Plots received single pulses of nitrate (0.5L of 300 mM)

mm

ol N

2O

m-2

h-1

-60

-50

-40

-30

-20

-10

0

10

20

30m

mo

lC

O2

eq

. m

-2d

ay

-1

= CO2 sequestration*

= equivalent of N2O

= equivalent of N2O and CH4

Nitrate added Control

CO2 eq. for N2O CO2 eq. for CH4

and N2O

Average CO2 sequestration in wetlands*

N2O CH4 and N2O

*Source: http://cdiac.ornl.gov/SOCCR Moseman-Valtierra et al. 2011. Atmospheric Environment

…but what happens over longer time scales?

?

Two Long-term Experiments

Plum Island Sippewissett

7-8 year fertilization at Plum Island (T.I.D.E.S.)

70 mM NO3-

15X background

Map from Johnson and Fleeger et al. 2009

During a transient doubling of fertilization…

= Pre- fertilization pulse (7/12)

= Pulse of 30X Fertilization (7/17)

= Reference marsh (Unfertilized)

mm

ol N

2O

m-2

h-1

-0.5

0.5

1.5

= Post- fertilization Pulse (7/19) Pre-

Pulse

Post-

Why were N2O fluxes so low?

Porewater nutrient levels

0102030405060

1 2 3C

on

cetr

atio

n (

uM

)

average Nitrate (uM)

average Ammonium (uM)

(mM)

Creek water nutrient levels

During pulse: Nitrate on Flood tide averaged 112 mM

(13 mM NO3- during flux measurements)

Ammonium was less than 18 mM

Pre- Pulse Post-

(mM)

(mM

)

N2O fluxes reflect porewater nitrate concentrations

0 0

2.1

0.2 0

0

0.5

1

1.5

2

2.5

1 2 3 4 5mm

ol N

2O

m-2

h-1

Plot Number

0.01.02.03.04.05.06.07.08.09.0

1 2 3 4 5

nit

rite

+nit

rate

(mM

)

Plot Number

Direction of N2O flux reverses when the fertilization ceases

-0.5

0.5

1.5

-10

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

um

ol N

2O

m-2

h-1

mm

ol N

2O

m-2

h-1

mm

ol N

2O

m-2

h-1

2010 2011

Why did results of long- and short- term experiments differ?

-20

-15

-10

-5

0

5

0 50 100 150 200

um

ol N

2O

m-2

h-1

nitrite + nitrate (mM)

Long-term N short term N

Conclusions

• Long term (7-8 year) fertilization did not affect N2O or CH4 fluxes…but nutrient concentrations were not very high and not constant.

• Marshes may be resilient! N reductions may prevent or limit GHG emissions.

Plum Island

Map: Fox et al. 2012 Estuaries and Coasts

XF=Extra-high fertilization (7.56 g N m-2 wk-1)

HF=high fertilization (2.52 g N m-2 wk-1) LF= low fertilization (0.86 g N m-2 wk-1) Control= no fertilization

41-42 year fertilization at Sippewissett Marsh

N2O fluxes display spatial variability

LF, 0

-20

-15

-10

-5

0

5

mm

ol

N2O

m-2

h

-1

2010

= XHF, Extra-high fertilization (7.56 g N m-2 wk-1)

= XF (7.56 g N m-2 week-1)

= HF (2.52 g N m-2 week-1)

= LF (0.86 g N m-2 week-1)

= Control ( 0 g N m-2 week-1)

No significant differences between treatments

Nearly all N2O fluxes completely reverse directions between 2 years

-5

0

5

mm

ol

N2O

m-2

h -1

LF, 0

-20

-15

-10

-5

0

5

mm

ol

N2O

m-2

h

-1

2010 2011

Why such a difference between years?

• One possibility is temperature:

No difference over time in: Salinity, pH, Oxidation reduction

23

27.6

0

10

20

30

Tem

pe

ratu

re (

de

g. C

)

2010 2011

Highest N2O flux in plot with highest pore water ammonium

-20

-15

-10

-5

0

5

10

0 100 200 300 400

um

ol N

2O

m-2

h-1

ammonium (mM)

Sippewissett Plum Island short term N Plum Island

Combined data from 2010 and 2011 in both experiments

0

10

20

30

40

50

60

um

ol C

H4

m-2

h-1

0 0

-10

0

10

20

30

40

50

um

ol C

H4

m-2

h-1

Highest methane fluxes in plots with high fertilization (HF)

2010 2011

F3=8.89, p=0.05

12

22

0

55

0 0 0 0 0

20

40

60

1 2 3 4 5 6 7 8mm

ol C

H4 m

-2 h

-1

Plot Number

…but high variability in 2011 = XF (7.56 g N m-2 week-1)

= HF (2.52 g N m-2 week-1)

= LF (0.86 g N m-2 week-1)

= Control ( 0 g N m-2 week-1)

mm

ol C

H4 m

-2 h

-1

mm

ol C

H4 m

-2 h

-1

-40

-20

0

20

40

60

80

100

120

-200 -100 0 100 200 300 400

mm

ol C

H4 m

-2 h

-1

Oxidation-Reduction Potential (Eh)

XHF HF LF control

= XF (7.56 g N m-2 week-1)

= HF (2.52 g N m-2 week-1)

= LF (0.86 g N m-2 week-1)

= Control ( 0 g N m-2 week-1)

Soil oxidation-reduction potential

Why were methane fluxes not as large in plots with the highest fertilization (XF)?

Several questions remain

Conclusions

• Long term fertilization (40 years) did not significantly change N2O fluxes (in one of the plant zones) -Reversal of the direction of N2O fluxes may have

been related to temperature

(Does warming lead to higher emissions?)

• CH4 fluxes were significantly higher in the High Fertilization plot (in one of 2 years)

-2

-1

0

1

2

3

4

5

6

7

8

9

um

ol N

2O

m-2

h-1

2010-2011

Large N2O fluxes of N-enriched marshes

Marshes without N enrichment Sippewissett

Plum Island (Sweeney)

Short term N pulse at Plum Island

Chronically enriched riparian buffer (Hefting et al. 2003)

(Yu et al. 2007 Hopfensperger et al. 2009)

light

dark

XF

XF 2011

31

-4.21

9 29

690

1247

-200

0

200

400

600

800

1000

1200

1400

um

ol C

H4

m-2

h-1

Fresh riparian Hopfensperger et al. 2009

Boreal salt marsh

Brackish marsh (DeLaune et al. 1983)

Magenheimer et al. 1996

CH4 fluxes are not large relative to other wetlands

Sippewissett

Plum Island (Sweeney)

Fresh Phragmites marsh

Flury et al. 2010

Future directions: The “Real” N experiments Studying relationships of GHG fluxes to

plant zones with in situ analyzers

NERR Science Collaborative: K. Kroeger, J. Tang, O. Abdul-Aziz, N. Ganju, A. Leschen, T. Surgeon-Rogers, S. Emmett-Mattox, I.Emmer, S. Crooks, P. Megonigal, T. Walker, C. Weidman

Colors indicate distinct zones in Sage Lot Pond (Waquoit Bay)

Waquoit Bay, MA

Narragansett Bay, RI Testing potential interactions of N loading and warming

Collaborator: B. Govenar

Acknowledgements

• Students: Rosalinda Gonzalez (Woods Hole Partnership Education Program), Kelsey Fisher, David McChesney

(Boston College)

• Collaborators : Dr. Linda Deegan, Dr. Ivan Valiela,

Dr. Jianwu Tang (Marine Biological Laboratory),

Dr. Breea Govenar (Rhode Island College)

Field support: (USGS)Wally Brooks, Adrian Green, Sandy Baldwin; Kate Morkeski, David Johnson (MBL)

Nutrient Analyses: Linda Green (URI), Paul Henderson (WHOI)

• Funding: USGS Mendenhall Post-Doctoral Research Fellowship

USDA Agricultural Experiment Station and University of Rhode Island (start-up funding)

National Estuarine Research Reserve System Science Collaborative 2012-2014

Rhode Island Research Alliance Collaborative Grant 2012